The chief of South Korea’s SK Group has argued that the AI industry is not in a bubble, stressing that long-term growth prospects remain robust even as investment pours into the sector. At a recent Asia-Pacific industry event on December 5, 2025, he warned that AI-related stocks could still face a correction as market enthusiasm cools after rapid gains. His assessment draws a clear line between the sector’s underlying technological progress and the short-term valuation pressures building in public markets.
SK Group’s Leadership on AI Outlook
SK Group chairman Choi Tae-won used his appearance at the Asia-Pacific gathering to signal that he sees the AI sector as structurally sound, citing the scale and persistence of global capital flowing into data infrastructure, chips and software as evidence of genuine demand rather than speculative excess. According to his comments, the breadth of investment across regions and industries suggests that AI is becoming a foundational technology, not a passing trend, which is why he argued that the industry itself is not in a bubble. For policymakers and corporate leaders weighing long-term commitments to AI, his stance implies that backing core capabilities such as semiconductors and cloud capacity remains a defensible strategy even if market sentiment turns.
At the same time, Choi acknowledged that the financial side of the story looks more fragile, noting that AI-linked equities have surged on expectations that may be difficult to meet in the near term. In remarks highlighted by his warning that the AI industry is not in a bubble but stocks could see a correction, he suggested that valuations in some parts of the market have run ahead of realistic earnings trajectories. For investors and pension funds that have piled into AI names, that distinction between solid industrial momentum and stretched share prices raises the risk of short-term losses even as the long-range thesis remains intact.
Distinguishing AI Fundamentals from Stock Valuations
Choi’s argument rests heavily on the underlying technology stack that supports AI, from advanced chips to large-scale data centers and integrated software platforms. He pointed to the way major players are knitting together hardware and software, and he underscored that SK Group is actively expanding its own semiconductor and data center capabilities to support this growth trajectory. By emphasizing these concrete build-outs, he framed AI as an infrastructure story, where capital is being deployed into physical assets and mission-critical systems that enterprises in manufacturing, healthcare and logistics are already using in production. For corporate IT buyers and industrial operators, that focus on real deployments rather than prototypes suggests that AI spending is likely to persist even if equity markets wobble.
In parallel remarks reported by regional outlets, Choi reiterated that while the AI industry shows no signs of a bubble, overvalued equities might adjust to reflect more realistic pricing. He contrasted the current environment with past tech bubbles, arguing that today’s AI systems are embedded in tangible applications such as predictive maintenance in factories, clinical decision support in hospitals and automated quality control in electronics production. For portfolio managers, that comparison matters because it suggests that even if some high-flying names fall sharply, the broader ecosystem of suppliers, integrators and users is less likely to experience a systemic collapse similar to earlier boom-and-bust cycles.
Mechanics of a Potential Stock Correction
Choi’s caution centers on the mechanics of a standard market correction, which he framed as a likely scenario for AI-linked stocks after their 2025 rally. In equity markets, a correction typically refers to a decline of 10 percent or more from recent peaks, and he indicated that such a pullback in AI names would primarily serve to remove excesses built up during the surge in enthusiasm around generative AI tools. By tying his comments to that threshold, he implicitly signaled that a double-digit drop in AI indices or leading chipmakers should be interpreted as a normalization rather than a sign that the AI growth story has failed. For retail investors who entered the sector late in the rally, that framing could help distinguish between painful but temporary mark-to-market losses and a fundamental break in the investment case.
He also pointed to macro and behavioral factors that could trigger or deepen such a correction, including rising interest rates and profit-taking by institutional investors that have enjoyed substantial paper gains. In his updated assessment, Choi noted that the sector’s 2025 run-up was driven in part by hype around generative AI tools, which encouraged momentum trading and concentrated flows into a narrow group of perceived winners. His more recent comments, which stress that a correction is likely but would not derail the underlying AI momentum established through 2025, suggest that SK Group is preparing for volatility in its own share price and in the valuations of its partners. For corporate treasurers and CFOs, that expectation of choppier markets may influence decisions on buybacks, capital raising and the timing of major AI-related investments.
Expert Insights on Healthy Market Corrections
Market strategists have long argued that corrections can play a constructive role in equity markets, and Choi’s remarks align with that view by framing a potential AI pullback as a reset rather than a crisis. Analysts interviewed in a detailed explainer on what constitutes a healthy correction in stocks describe these episodes as natural phases that weed out weaker performers and speculative excess. They note that when prices fall 10 to 20 percent after a period of over-enthusiasm, capital often rotates toward companies with stronger balance sheets, clearer earnings visibility and more defensible competitive positions. For AI, that could mean that firms with proven products, recurring revenue and diversified customer bases emerge stronger, while early-stage or marginal players struggle to raise funds on favorable terms.
Experts also emphasize that healthy corrections tend to follow periods when investors extrapolate short-term growth rates too far into the future, a pattern that closely mirrors the exuberance around generative AI models and infrastructure providers in 2025. In their view, a pullback that trims valuations without triggering widespread credit stress can ultimately foster long-term stability by re-anchoring expectations to achievable performance metrics. Choi’s December 5, 2025, comments, which represent an evolution from earlier expert consensus that largely downplayed downside risks in the AI boom, fit squarely within that framework. For regulators and central banks monitoring financial stability, the convergence between corporate leaders and market analysts on the value of a controlled correction may reduce the likelihood of abrupt policy interventions if AI stocks slide.
Shifting Tone From Early 2025 Optimism
Choi’s latest remarks also mark a notable shift from the tone that dominated SK Group’s public statements earlier in 2025, when the focus was largely on explosive stock surges and the transformative potential of AI. At that stage, corporate messaging tended to highlight upside scenarios, with less attention paid to the possibility that valuations could overshoot even if the technology delivered on its promises. By now incorporating explicit warnings about potential volatility tied to investor sentiment, Choi is signaling that SK Group sees a more complex landscape in which operational execution and capital market dynamics must be managed in parallel. For shareholders, that evolution in tone may be interpreted as a move toward more balanced guidance, which can help temper unrealistic expectations and reduce the shock if prices retrace.
The updated stance also reflects how quickly narratives around AI can change as markets digest new information about adoption rates, regulatory responses and competitive pressures. Earlier in the year, many commentaries framed the AI surge as a one-way trade, with limited discussion of how rising borrowing costs or shifts in risk appetite might affect valuations. Choi’s current view, which stresses that fundamentals remain strong even if prices moderate, implicitly acknowledges that AI companies will need to navigate a more demanding environment in which investors scrutinize profitability, unit economics and capital intensity more closely. For boards and executive teams across the sector, that shift underscores the importance of aligning investor communications with realistic roadmaps rather than relying on broad promises of disruption.
Implications for Investors and the Broader Tech Ecosystem
For institutional and retail investors, the combination of Choi’s confidence in AI’s industrial trajectory and his caution on stock prices presents a nuanced roadmap for positioning portfolios. His argument suggests that exposure to AI should increasingly be built around companies with durable competitive advantages in semiconductors, cloud infrastructure and applied software, rather than purely momentum-driven plays. Investors may also need to prepare for higher volatility, using tools such as staggered entry points, diversification across the AI value chain and more rigorous valuation screens to manage risk. In practical terms, that could mean balancing holdings in headline AI names with stakes in suppliers of memory chips, networking equipment and specialized software that benefit from AI adoption but trade at less extreme multiples.
Beyond capital markets, Choi’s comments carry implications for the broader tech ecosystem, including startups, regulators and end users. Startups that have relied on lofty public-market comparables to justify high private valuations may face tougher fundraising conditions if listed AI stocks correct, which could accelerate consolidation or force business model pivots. Regulators, meanwhile, may interpret the prospect of a controlled correction as evidence that markets are capable of self-adjustment, potentially reducing pressure for direct intervention in AI-related valuations while keeping the focus on issues such as data governance and competition. For enterprises deploying AI in sectors like manufacturing and healthcare, the key takeaway is that while vendor valuations may fluctuate, the strategic imperative to integrate AI into operations remains intact, reinforcing the need for long-term planning that looks beyond short-term market swings.
